Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Seeed-Projects/Seeed-Jetson-DevelopTool --skill rebot-arm-graspgit clone --depth 1 https://github.com/Seeed-Projects/Seeed-Jetson-DevelopToolWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/seeed-projects/seeed-jetson-developtool/rebot-arm-grasp)<a href="https://agentmods.dev/skills/seeed-projects/seeed-jetson-developtool/rebot-arm-grasp"><img src="https://agentmods.dev/badge/skills/seeed-projects/seeed-jetson-developtool/rebot-arm-grasp/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/seeed-projects/seeed-jetson-developtool/rebot-arm-grasp"><img src="https://agentmods.dev/badge/skills/seeed-projects/seeed-jetson-developtool/rebot-arm-grasp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00065 | $0.02064 |
| Opus 5.5 | $0.00026 | $0.00826 |
| Sonnet 5.5 | $0.00013 | $0.00413 |
| Haiku 4.5 | $0.00006 | $0.00206 |
Grade A, and why
rebot-arm-grasp scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
reBot 物体抓取
让 Codex 作为决策 agent,使用一个持续运行的 CASCADE MCP 会话完成观察 → 定位 → 抓取/搬放 → 结果核验。关节控制、IK、速度限制、障碍物和接触检查继续由 CASCADE 执行。此 skill 是操作层,运行时依赖完整 CASCADE 源码、机器人资产和现场配置;单独安装 skill 不会安装驱动或获得硬件验收。
入口与连接
- 已有 CASCADE MCP:先读取
world_state,确认当前实际 arm、camera 和能力;不要创建第二个总线控制进程。工具名前缀取决于 MCP 注册名称,以实际tools/list为准。 - 尚未连接:读 setup.md,运行
scripts/runtime.py check。默认是纯 mock。check只检查本地文件/依赖,不访问设备,也不代表实机 READY。 - 实机接入前读 hardware.md。本包实机启动器支持
type: rebot_rs的 B601 RobStride SDK 路径;rebot_rs_mb、DM 电机和其他机械臂需要各自验证,不自动切换。 - 用
scripts/runtime.py mcp-config生成配置,serve持续提供原生 MCP。启动器将实机置于软件 STOP;用户当前抓取指令已授权实机运动且现场检查通过时,可以按实际 schema 调用reset_stop。历史急停、接触故障或操作员禁用的 reset 必须先解决原因,不自动清除。 - 本包没有通用实机
call/批量运动脚本。实机调用通过已注册 MCP 会话进行;smoke_test.py被固定为 mock。
命令示例(SKILL 是本技能所在目录,REPO 是目标 CASCADE checkout 的绝对路径):
python3 "$SKILL/scripts/runtime.py" check --repo "$REPO"
python3 "$SKILL/scripts/runtime.py" smoke --repo "$REPO"
python3 "$SKILL/scripts/runtime.py" check --repo "$REPO" \
--mode hardware --arm rebot_rs --cameras YOUR_CALIBRATED_CAMERA
每次任务
- 读
world_state,检查stopped、tracked_holding、gripper_contents、backends、capabilities、tools_withheld和操作员隐藏项。MCP 的 tracked_holding 是 session 记录,gripper_contents 通常是 not_measured;其他界面可能称 holding/estopped。待接触/释放、可能仍有负载、标定失效或地图错误时,进入 recovery.md,不要通过改限制恢复运动。 - 调用
get_observation或describe_scene获取当前图像。camera_snapshot可选择 exact camera name。确认目标身份、所在工作区域及放置目的地;同类物体不唯一时用localize_object(label, spatial_hint),仍不明确就澄清。自然语言可用中文;参数保留工具要求的名称/枚举,物体 label 按实际图像和识别结果表达。 - 实机用
localize_object和preview_grasp检查目标位置、尺寸、拟议夹爪宽度/方向。位置是机器人基座坐标、米;关节角是弧度。预览仅是候选计划,不证明完整运动可执行,实际抓取仍要重新感知并经过 IK/安全检查。configured zone 是配置来源,不证明它可见或为空。 - 按用户意图选择一个工具:
- “抓住/拿起来并保持”:
grasp_object({"label": "red cube", "material": "rigid"})。完成后保留会话及负载;不追加放下、开爪或断电。 - “抓取并放到 …”:优先
pick_and_place({"object": "red cube", "destination": "已确认的目的地"});仅用户接受当前配置的 drop zone 时才省略 destination。此工具包含有界重试和返回 home,检查return_home,不重复调用move_home。 - 已持物:先核实持物身份和负载状态,再用
place_on_object,或用本次观测/现场确认的基座坐标调用place_at。不重新抓取另一物体。
- “抓住/拿起来并保持”:
- 查看动作结果中的
ok、execution_ok(若有)、stage、outcome、postcondition、return_home;再读verify_last_action和新观测。verify_last_action读取已保存 verdict,不重新测量,也不能升级缺失证据。多步任务在下一动作前用task_memory检查已完成步骤。 - 报告目标、实际动作、结果状态和证据来源,并给出 trace/run 路径。
postcondition.status为confirmed才称该后置条件独立验证成功;refuted是失败;unverified明确说缺乏独立证据。tracked_holding、belief 更新、模型置信度、ok: true、端口连通都不能单独证明物理成功。Mock 的静态图像通常 abstain,不能声称物体真的移动。
What ships with it
13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 157 B
- LICENSE 1.0 KB
- references/hardware.md 4.4 KB
- references/recovery.md 2.8 KB
- references/setup.md 4.1 KB
- references/source-manifest.json 3.6 KB
- references/tools-and-results.md 4.0 KB
- references/validation-result.json 4.6 KB
- references/validation.md 3.4 KB
- scripts/mock_fixture_server.py 1.3 KB runs code
- scripts/runtime.py 17 KB runs code
- scripts/smoke_test.py 7.4 KB runs code
- scripts/test_runtime.py 8.8 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 55 lines · 65 tokens per session scan A 17e64f103f52
rebot-arm-grasp is a skill published in the GitHub repository Seeed-Projects/Seeed-Jetson-DevelopTool (56 stars, last pushed yesterday), licensed MIT. It adds 65 tokens to every session and 2,064 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-24.
Other skills, from other repositories
gke-compute-classes
Configures, optimizes, and troubleshoots GKE ComputeClasses. Use when configuring Spot VMs with on-demand fallback, targeting specific accelerators (GPUs/TPUs) or machine families, restricting ComputeClass access, or debugging pending pods related to node pool auto-creation. Do not use for cluster-level Node Auto…
jetson-diagnostic
Read-only Jetson health snapshot for identity, memory, GPU, thermal, power, storage, services, and top processes.
holoscan-install-wheel
Install Holoscan SDK 5.x CUDA 13 Python wheels in an isolated environment, choosing the full stack or version-matched components.
offensive-z-wave
Z-Wave attack methodology — sniffing with Z-Force / EZ-Wave / RTL-SDR + ZniffMobile, S0 (legacy) network-key derivation flaw and key reuse, S2 (modern) ECDH commissioning analysis, replay/injection on unauthenticated nodes, default-key brute-force on test deployments, and home-automation hub pivots. Use when targeting…
doca-socket-relay
Use this skill when the operator is driving the DOCA Socket Relay to bridge a socket-oriented host application onto a BlueField DPU peer without rewriting it — picking the deployment shape (in-process, sidecar, or BlueField service container), configuring the host-side socket and the DPU-side forwarding endpoint…
hsb-flash
Flash the FPGA on an HSB board connected to an NVIDIA devkit. Supports HSB Lattice boards (FPGA versions 2407, 2412, 2507, 2510) and Leopard Imaging VB1940 "all-in-one" cameras (FPGA versions 2507, 2510). Uses release-specific YAML manifests and board-type-specific program commands. Lattice and VB1940 commands must…